A Light-Inducible Strain for Genome-Wide Histone Turnover Profiling in Neurospora crassa

Abstract

In chromatin, nucleosomes are composed of ∼146 bp of DNA wrapped around a histone octamer, and are highly dynamic structures subject to remodeling and exchange. Histone turnover has previously been implicated in various processes including the regulation of chromatin accessibility, segregation of chromatin domains, and dilution of histone marks. Histones in different chromatin environments may turnover at different rates, possibly with functional consequences. Neurospora crassa sports a chromatin environment that is more similar to that of higher eukaryotes than yeasts, which have been utilized in the past to explore histone exchange. We constructed a simple light-inducible system to profile histone exchange in N. crassa on a 3xFLAG-tagged histone H3 under the control of the rapidly inducible vvd promoter. After induction with blue light, incorporation of tagged H3 into chromatin occurred within 20 min. Previous studies of histone turnover involved considerably longer incubation periods and relied on a potentially disruptive change of medium for induction. We used this reporter to explore replication-independent histone turnover at genes and examine changes in histone turnover at heterochromatin domains in different heterochromatin mutant strains. In euchromatin, H3-3xFLAG patterns were almost indistinguishable from that observed in wild-type in all mutant backgrounds tested, suggesting that loss of heterochromatin machinery has little effect on histone turnover in euchromatin. However, turnover at heterochromatin domains increased with loss of trimethylation of lysine 9 of histone H3 or HP1, but did not depend on DNA methylation. Our reporter strain provides a simple yet powerful tool to assess histone exchange across multiple chromatin contexts.

THE eukaryotic genome is packaged in the nucleus by wrapping DNA around histone octamers, forming the basic organizational unit called the nucleosome. Nucleosomes form the basis of the higher-order organization of the genome, which can impact the function of underlying genetic elements (van Steensel 2011). The positions of key nucleosomes within chromatin are subject to regulation, and underlying histones are subject to exchange (Venkatesh and Workman 2015). Histone exchange can be independent of DNA replication and, in at least in some cases, levels of histone turnover appear to correlate with levels of gene expression (Venkatesh and Workman 2015). High levels of turnover are thought to allow for greater accessibility to DNA for transcriptional machinery such as transcription factors and RNA polymerases, and may play a role in the regulation and distribution of histone marks, and histone variants (Venkatesh and Workman 2015). Histone turnover may also play a role as boundary elements to contain heterochromatin spreading (Allshire and Madhani 2018). Heterochromatin is generally thought of as a compacted form of chromatin, rendering its DNA relatively inaccessible to transcriptional factors and suppressing the activity of selfish genetic elements (Allshire and Madhani 2018). Consistent with this view, heterochromatin has been found to be refractory toward histone exchange, and disruption of heterochromatin leads to increased turnover in fission yeast (Choi et al. 2005; Aygün et al. 2013). Potentially, suppression of histone turnover could be a general feature of heterochromatin, which might contribute to epigenetic inheritance and other functions of this genome compartment.

In metazoans, replication-independent histone turnover, such as during transcription, results in replacement of hH3.1, which deposited during DNA replication, with a closely related histone variant, hH3.3. This exchange is thought to help expose DNA-binding sites and is apparently crucial for development (Venkatesh and Workman 2015). Investigations of histone turnover have been performed in the yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe, which offer a simpler complement of histone proteins; yeasts have only one hH3 isoform, a homolog of hH3.3 (Venkatesh and Workman 2015). Like yeasts, the filamentous fungus Neurospora crassa possesses only one histone H3 isoform, which is also homologous to metazoan hH3.3. However, N. crassa possesses chromatin features more similar to those in higher eukaryotes. For example, like higher organisms, N. crassa possesses two functionally distinct methyltransferases for lysine 36 on histone H3 (H3K36), SET-2 and ASH1, while yeasts possess just SET-2. SET-2 is responsible for nearly all trimethylated H3K36 in N. crassa and is implicated in regulating histone exchange in the wake of RNA polymerase II during its passage over gene bodies; ASH1-deposited dimethylated H3K36 appears to mark transcriptionally silent genes (Bicocca et al. 2018). N. crassa also sports DNA methylation at constitutive heterochromatin as well as the facultative heterochromatin mark, methylation of lysine 27 of histone H3 (H3K27me), both of which are absent from yeasts (Aramayo and Selker 2013).

Here, we report the construction and validation of a histone turnover reporter strain of N. crassa. Our strain utilizes 3xFLAG-tagged histone H3 under the control of a light-inducible promoter. Inducible tagged histones have been previously used to assay histone exchange in other models, but required the addition of an inducing agent—often through a change of growth medium—and a lengthy incubation period, typically on the order of hours (Ahmad and Henikoff 2002; Choi et al. 2005; Dion et al. 2007; Rufiange et al. 2007; Aygün et al. 2013; Kraushaar et al. 2013). Our use of light as the inducing signal is less disruptive than a change of medium and provides greater control and more rapid induction. We used our system to explore possible effects on histone turnover of mutants defective in heterochromatin machinery.

Materials and Methods

N. crassa strains and molecular analyses

Strains are listed in Supplemental Material, Table S1 and were grown, crossed, and maintained according to standard procedures (Davis 2000). Primers used are listed in Table S2. DNA isolation and western blotting were performed as previously described (Honda and Selker 2008). The following antibodies were used in western blot analyses: anti-FLAG-conjugated HRP (FLAG-HRP) (A8592; Sigma [Sigma Chemical], St. Louis, MO) and anti-phosphoglycerate kinase 1 (PGK1) (ab113687; Abcam). Chemiluminescence from treatment with SuperSignal West Pico Substrate (34080; Thermo Fisher Scientific) was used for anti-FLAG-HRP. Fluorescence of IRDye 680RD goat anti-rabbit (926-68071; Licor) was used for anti-PGK1. Both were imaged using a LI-COR Odyssey Fc imaging system.

Generation of P_vvd::hH3-3xFLAG::his-3⁺ reporter strain

The promoter region of the vvd gene was amplified by PCR using wild-type genomic DNA as template using primers VVD-3000F_EcoRI\NotI and VVD-R1_XbaI containing NotI and XbaI sites, respectively. The PCR product and pCCG::C-Gly::3xFLAG (Honda and Selker 2009) were digested with NotI and XbaI, and ligated together to create pVVD::C-Gly::3xFLAG. pVVD::C-Gly::3xFLAG and pIDTSMART-AMP containing N. crassa codon-optimized hH3 (New England Biolabs, Beverly, MA) were digested with XbaI and PacI, and ligated together to yield pVVD:hH3::C-Gly::3xFLAG. The ligation product was verified by restriction analysis and Sanger sequencing. pVVD::hH3::C-Gly::3xFLAG was then linearized and transformed into strains N2930 and N3012. The resulting his-3⁺ strains (N6049 and N6054) were verified by Southern hybridization.

Histone exchange profiling and analysis

Fresh conidia (10⁶/ml) were inoculated in Vogel’s medium N with 1.5% sucrose, and grown for 18 hr at 32° with shaking (150 rpm) in a dark room. Low-intensity red light was used for working during all culturing steps to prevent light stimulation (Aronson et al. 1994). To block DNA replication, cultures were spiked with hydroxyurea (HU) to a final concentration of 100 mM, and incubated for 3 hr (Martegani et al. 1981; Sachs et al. 1997). We affixed five, 2-ft long strips of blue light light-emitting diode (LED) lights (EL-BVRIB12V, power supply: EL-12VADPT, dimmer switch: EL-SC12DIM, and DC 5-way splitter: EL-SJDCSPLIT; Elemental LED) to the bottom of a New Brunswick G25 incubator lid, which placed them ∼16 cm from the top of the growth medium of cultures. To induce H3-3xFLAG expression, the LED strips were activated to 100% power and cultures were exposed to blue light (465 nm; 30 µmol photons/m² per sec) for 2 min. Cultures were incubated for 30 min (unless noted otherwise) to allow incorporation of H3-3xFLAG into chromatin and then immediately harvested, washed with PBS buffer, and transferred to 125-ml Erlenmeyer flasks with 10 ml PBS buffer. For chromatin fixation, formaldehyde was added to a final concentration of 0.5% and incubated for 30 min at room temperature. The cross-linking reaction was quenched with glycine (final concentration of 0.2 M) and then incubated for an additional 5 min at room temperature. Tissue was disrupted by sonication, and chromatin was sheared with a Bioruptor (Diagenode) for 20 min with 30-sec on/off intervals at high power. Subsequent chromatin immunoprecipitation (ChIP) was performed as previously described (Tamaru et al. 2003) using anti-FLAG-conjugated agarose beads (A2220; Millipore, Bedford, MA) and anti-hH3 (Ab1791; Abcam).

For ChIP-quantitative PCR (qPCR) analyses, independent experimental replicates were performed in triplicate using PerfeCTa SYBR Green Fastmix ROX (Quantabio) with the listed primers (Table S2), and analyzed using a StepOnePlus Real-Time PCR system (Life Technologies). Relative enrichment was determined by calculating enrichment as a percentage of the total input.

H3-3xFLAG ChIP samples were prepared for sequencing as previously described (Jamieson et al. 2016). Sequencing was performed using an Illumina HiSeq 4000 system with single-end 75-nt or 100-nt reads. Sequencing data were processed and analyzed using the Galaxy platform (https://usegalaxy.org/) (Afgan et al. 2016). Sequences were aligned to the corrected N. crassa OR74A (NC12) genome (Galazka et al. 2016) using Bowtie2 (Langmead and Salzberg 2012). For visualization, BedGraph files were generated from mapped read data using Hypergeometric Optimization of Motif EnRichment (HOMER) (Heinz et al. 2010), converted into bigWig files using the Wig/BedGraph-to-bigWig program via Galaxy, and visualized with the Integrative Genomics Viewer (Thorvaldsdóttir et al. 2013). Meta-analyses were carried out using the computeMatrix and plotProfile programs from deepTools (Ramírez et al. 2016) via Galaxy.

Light characterization

The light spectral range was measured using an ILT350 chroma meter (International Light Technologies) and intensity was measured using an MQ-200 quanta meter (Apogee Instruments).

Data availability

The histone turnover reporter strains used in this study are available on request from the Fungal Genetics Stock Center. Complete ChIP-sequencing (ChIP-seq) and bisulfite sequencing files, ChIP-seq intensity values, and gene expression quartile lists and constitutive heterochromatin region coordinate files have been deposited in the National Center for Biotechnology Information’s Gene Expression Omnibus (GEO; http://ncbi.nlm.nih.gov/geo), and are accessible through GEO Series accession number GSE143608 and as part of a previously reported series (GSE81129). Supplemental material available at figshare: https://doi.org/10.25386/genetics.12032664

Results and Discussion

Expression of hH3-3xFLAG under control of the vvd promoter

To assay histone turnover genome-wide in N. crassa, we constructed a strain expressing 3xFLAG-tagged histone H3 at the his-3 locus, regulated by the vvd promoter (Figure 1A). N. crassa possesses a well-characterized circadian rhythm, and > 5% of its genes are expressed in response to light exposure (Chen and Loros 2009). The vvd promoter allows for light-inducible expression of genes; in dark conditions, vvd is weakly expressed but is rapidly induced by up to 300-fold within minutes of light exposure (Heintzen et al. 2001; Elvin et al. 2005). Indeed, the vvd promoter has already been successfully utilized for the inducible and tunable control of gene expression (Hurley et al. 2012). Since complete histone eviction requires removal of H2A-H2B dimers, and the H3-H4 tetramer is incorporated into chromatin prior to other histones, the incorporation of H3-3xFLAG should represent the turnover of entire nucleosomes (Luger et al. 2012).

Figure 1.

A simple light-inducible system for assessing replication-independent histone turnover in N. crassa. (A) Schematic overview of the light-inducible histone H3-3xFLAG reporter strain. (B) Expression time course analysis of H3-3xFLAG induction. H3-3xFLAG was induced with a 2-min pulse of blue light and H3-3xFLAG levels were examined at 20-minute intervals by immunoblotting. Uninduced H3-3xFLAG levels are indicated by “DD.” Phosphoglycerate kinase 1 (PGK1) levels are included for each time point as a loading control.

To examine the kinetics of light-inducible expression of H3-3xFLAG and to characterize its expression dynamics postinduction, we grew cultures of the reporter strain in complete darkness for 18 hr, exposed them to a short period (2 min) of blue light, and assessed H3-3xFLAG expression by immunoblotting in a time course experiment (Figure 1B). Consistent with previous characterization of gene expression controlled by the vvd promoter, we observed a low basal level of H3-3xFLAG expression at the preinduction time point (Figure 1B, uninduced H3-3xFLAG levels). By 20 min, there was a marked accumulation of H3-3xFLAG, which appeared to culminate by 40 min (Figure 1B). This level of H3-3xFLAG persisted for at least 2 hr after induction. Because slants used to inoculate the dark cultures were grown in light, P_vvd:hH3-3xFLAG may have been active at the start of the incubation period. The low level of H3-3xFLAG observed in the uninduced strains suggests that high levels brought on by light induction were effectively reduced to this low basal level within 18 hr. Conceivably, longer growth periods in the dark might reduce background levels further.

H3-3xFLAG is incorporated into chromatin

To verify incorporation of light-induced H3-3xFLAG into chromatin, we immunoprecipitated formaldehyde cross-linked FLAG-tagged histones after various periods of incubation to allow for histone incorporation, and tested for association with DNA by qPCR. We expected that longer incorporation times would give greater enrichment of immunoprecipitated chromatin. To assess whether FLAG-tagged histones were effectively incorporated in a replication-independent manner, we blocked DNA replication by treatment with HU. Cultures were spiked with 100 mM HU and incubated for 3 hr, a regimen previously shown to effectively arrest replication without compromising viability (Martegani et al. 1981; Sachs et al. 1997). Strains were then exposed to a 2-min light pulse to induce H3-3xFLAG expression and incubated for 20 min or 2 hr to allow for incorporation of FLAG-tagged histones before cross-linking, chromatin shearing, immunoprecipitation, and isolation of associated DNA (Figure 2).

Figure 2.

Flowchart for genome-wide profiling of replication-independent histone turnover in N. crassa. First, 5 ml of Vogel’s medium N supplemented with 1.5% sucrose is inoculated with 5.0 × 10⁶ conidia, and cultures are grown in complete darkness at 32° for 18 hr. To block DNA replication, cultures are treated with 100 mM HU and then induced with light 3 hr later. After 30 min to allow for incorporation of new H3-3xFLAG into chromatin, cultures are cross-linked, lysed, and the chromatin sheared. H3-3xFLAG is immunoprecipitated and the associated DNA isolated and analyzed by qPCR, or prepared for next-generation sequencing and sequenced. ChIP, chromatin immunoprecipitation; qPCR, quantitative PCR.

We performed qPCR using primers for different chromatin contexts: active genes (actin, fkr-5, and csr-1), constitutive heterochromatin (Cen IIIL, 8:A6, and 8:F10), and facultative heterochromatin (Tel VIIL). We observed similarly low levels of relative enrichment across all regions in the uninduced control. After only 20 min of incorporation after induction, enrichment increased at all regions, and further enrichment was found after 2 hr of incorporation (Figure 3A, left panel). These results indicate that light-induced H3-3xFLAG was readily incorporated into chromatin, and that longer incubation periods following induction resulted in increased incorporation of H3-3xFLAG.

Figure 3.

Histone turnover at representative genomic regions. (A) H3-3xFLAG incorporation into chromatin increases with incubation time. qPCR analysis of FLAG ChIP (left) and hH3 ChIP (right) after 20 min or 2 hr of incorporation, at different chromatin regions including: active genes (actin, fkr-5, and csr-1), constitutive heterochromatin (Cen IIIL, 8:A6, and 8:F10), and facultative heterochromatin (Tel VIIL). Control uninduced levels are indicated by “DD.” Data are presented as the mean and SD of three technical replicates. (B) Genome-wide profiling of histone turnover in N. crassa. Representative IGV tracks displaying H3-3xFLAG enrichment after 30 min of incorporation from replicate ChIP-sequencing experiments. Levels of DNA methylation (5mC), determined by bisulfite sequencing, are displayed to identify constitutive heterochromatin (Aramayo and Selker 2013). (C) Histone turnover metaplot of gene expression level. Genes were divided into expression quartiles based on wild-type expression levels, and H3-3xFLAG enrichment from two replicate experiments (denoted by solid and dotted lines) was aligned relative to their TSS and TES sites, and scaled to 3 kb (number of genes in each quartile: top 25%, 2433; 75–50%, 2432; and 50–25%, 2431; bottom 25%, 2433). H3-3xFLAG enrichment is displayed across the aligned gene bodies as well as 1-kb upstream regions for each quartile. 5mC, 5-methylcytosine ChIP, chromatin immunoprecipitation; IGV, Integrative Genomics Viewer; LG, linkage group; qPCR, quantitative PCR; rep, replicate; RPKM, reads per kilobase of transcript per million mapped reads; TES, transcription end site; TSS, transcription start site.

Considering the rapid induction and magnitude of H3-3xFLAG expression, we were concerned about the possibility that this would lead to abnormal levels of histones within chromatin. We therefore immunoprecipitated total hH3 after each incorporation period, and compared relative enrichment of associated DNA from each region across all incorporation time points by qPCR. For all regions examined, we found similar levels of enrichment for both incorporation time points as our uninduced control, suggesting that nucleosome concentration within chromatin was unchanged for at least 2 hr after induced expression of H3-3xFLAG (Figure 3A, right panel).

Replication-independent histone turnover in N. crassa

We examined replication-independent histone turnover genome-wide in N. crassa by performing next-generation sequencing on H3-3xFLAG-associated chromatin. We were concerned that a lengthy incorporation period might lead to saturation of FLAG-tagged histones within chromatin, resulting in enrichment values that might be more representative of histone occupancy than rates of histone turnover. To assess the incorporation of a “pulse” of labeled histones, we allowed only 30 min for incorporation (Figure 2). Based on our qPCR results, increasing amounts of H3-3xFLAG were still incorporated between 20 min and 2 hr after induction (Figure 3A). We reasoned that at 30 min, the size of any observed peaks of enrichment would reflect relative histone turnover rates rather than occupancy.

We sequenced DNA pulled down in two replicate turnover experiments and observed reproducible H3-3xFLAG enrichment patterns genome-wide (Figure 3B). The most prominent peaks were present over promoter regions of genes, which were typically adjacent to a region showing low turnover in the corresponding gene body, consistent with previous reports investigating replication-independent histone turnover in budding yeast (Figure 3B) (Dion et al. 2007; Rufiange et al. 2007; Kaplan et al. 2008; Gossett and Lieb 2012). Similarly, previous studies of replication-independent exchange of the histone H3 variant, hH3.3, in mouse and human cell cultures revealed high levels of turnover at promoter regions, 5′-UTRs, and the 3′ ends of genes (Jin et al. 2009; Goldberg et al. 2010; Kraushaar et al. 2013). There was a noticeable lack of histone turnover at constitutive heterochromatin domains, as previously observed in fission yeast (Choi et al. 2005; Aygün et al. 2013). Altogether, our findings with N. crassa support the general correlations between histone turnover and chromatin contexts found in other models.

Histone turnover in Neurospora genes

To characterize histone turnover in genes, we used previously generated RNA-sequencing data to divide all genes by expression level in wild-type cells into quartiles (Bicocca et al. 2018), aligned them relative to their putative transcriptional start and end sites, and averaged the relative signal across the promoter regions and gene bodies for each quartile. We found that each expression quartile possessed a distinct turnover profile (Figure 3C). Similar to observations in S. cerevisiae, Drosophila melanogaster, Caenorhabditis elegans, mouse, and human cells, increased levels of histone turnover correlated with higher relative expression levels (Ahmad and Henikoff 2002; Chow et al. 2005; Mito et al. 2005; Dion et al. 2007; Rufiange et al. 2007; T. Kaplan et al. 2008; Shivaswamy et al. 2008; Jin et al. 2009; Deal et al. 2010; Goldberg et al. 2010; Gossett and Lieb 2012). As in other model organisms, higher levels of histone turnover were found over promoter regions and in the 5′ and 3′ ends of genes (Mito et al. 2005; Dion et al. 2007; Rufiange et al. 2007; Jin et al. 2009; Ooi et al. 2009; Deal et al. 2010; Goldberg et al. 2010; Gossett and Lieb 2012) (Figure 3C). Notably, the interiors of gene bodies exhibited relatively low turnover, potentially due to SET-2-catalyzed H3K36me-mediated suppression of histone exchange (Venkatesh and Workman 2015). Consistent with this possibility, previous work on H3K36me in N. crassa revealed that SET-2-mediated H3K36me is mainly enriched over gene bodies, marking most (∼80%) N. crassa genes, although not those with low or no expression (Bicocca et al. 2018). The level of turnover within gene bodies was lower in the top three expression quartiles than the basal turnover found at genes within the lowest quartile, perhaps reflecting this mechanism, which is dependent on RNA polymerase II elongation (Venkatesh and Workman 2015).

The overall shapes of histone turnover profiles over genes in N. crassa appear to reflect an additional conserved feature of nucleosome organization in gene bodies of eukaryotes. Previous genome-wide studies uncovered nucleosome-free regions (NFRs) present immediately upstream of the transcriptional start sites in yeasts (Yuan et al. 2005; Albert et al. 2007; Whitehouse et al. 2007; Mavrich et al. 2008a; Kaplan et al. 2009), worms (Johnson et al. 2006; Valouev et al. 2008), flies (Mavrich et al. 2008b), medaka (Sasaki et al. 2009), and humans (Ozsolak et al. 2007; Schones et al. 2008). Expressed genes in N. crassa, represented by the top three expression quartile profiles, appear to organize nucleosomes at the 5′ ends of gene bodies as in other model systems. There are noticeable dips in H3-3xFLAG enrichment just upstream of the transcriptional start site, presumably due to the presence of NFRs (Figure 3C). Overall, it appears that the general associations between active transcription and histone turnover in N. crassa are similar to those found in other eukaryotes.

Histone turnover in N. crassa heterochromatin mutants

Heterochromatin is generally thought to be compact, largely transcriptionally silent chromatin that renders underlying DNA inaccessible to trans-acting factors (Allshire and Madhani 2018). Studies in S. pombe have shown heterochromatin to be refractory toward replication-independent histone turnover relative to transcriptionally active euchromatin (Choi et al. 2005; Aygün et al. 2013). Mutants deficient for heterochromatin-associated factors such as the histone deacetylase (HDAC) Clr3, the H3K9-specific methyltransferase Clr4, or the chromodomain proteins that bind this histone mark, Swi6 and Chp2, exhibit increases in histone turnover at heterochromatin domains, suggesting destabilization of heterochromatin structure (Choi et al. 2005; Aygün et al. 2013). Constitutive heterochromatin in N. crassa sports distinct histone marks and conserved features typical of this category of chromatin. Trimethylation of lysine 9 of histone H3 (H3K9me3) is catalyzed by the histone methyltransferase DIM-5 (Tamaru et al. 2003; Lewis et al. 2010). This mark is recognized by the chromodomain protein HP1 (Freitag et al. 2004), which functions as a platform to recruit the DNA methyltransferase DIM-2 to methylate associated cytosines (Honda and Selker 2008). HP1 also assists in directing the HDAC complex, HCHC, to heterochromatin, whose activity is dependent on its catalytic subunit, HDA-1 (Honda et al. 2012, 2016). To test how the loss of different heterochromatin factors may affect histone turnover in N. crassa, we built reporter strains in ∆dim-2, ∆dim-5, ∆hpo, and ∆hda-1 backgrounds, and assayed for H3-3xFLAG incorporation by next-generation sequencing.

In all mutant backgrounds tested, H3-3xFLAG patterns in euchromatin were almost indistinguishable from that observed in wild-type, suggesting that loss of heterochromatin machinery has little effect on histone turnover in euchromatin (Figure 4A). However, there were differences in heterochromatin domains. Heterochromatin domains in wild-type strains exhibited relatively low levels of histone exchange, but were found to be flanked by short regions of rapid turnover (Figure 4, A and B). It is of note that high levels of histone exchange were also observed at heterochromatin boundaries in S. pombe and D. melanogaster, and that histone turnover has been previously suggested to limit heterochromatin spreading (Dion et al. 2007; Deal et al. 2010; Aygün et al. 2013).

Figure 4.

Histone turnover at heterochromatin domains in mutants affecting heterochromatin. (A) Representative IGV tracks displaying H3-3xFLAG enrichment in indicated heterochromatin mutants. Constitutive heterochromatin is marked by DNA methylation (5mC) and highlighted in yellow. (B) Metaplot of histone turnover at heterochromatin domains in heterochromatin mutants. Constitutive heterochromatin domains were aligned and scaled to 2.5 kb. Replicate H3-3xFLAG enrichment profiles for WT (black) and the indicated heterochromatin mutant strains (red) are displayed over the scaled heterochromatin domains, and 1 kb up- and downstream from the domains. 5mC, 5-methylcytosine IGV, Integrative Genomics Viewer; LG, linkage group; RPKM, reads per kilobase of transcript per million mapped reads; WT, wild-type.

Histone turnover in heterochromatin in ∆dim-2 strains, which lack all DNA methylation, appeared largely normal in that turnover remained suppressed across heterochromatin domains. However, we observed a modest increase in turnover at the edges of heterochromatin domains (Figure 4, A and B). These results suggest that DNA methylation has little influence on histone turnover in N. crassa. Loss of DIM-5, and thus H3K9me3, or of the HDAC HDA-1, led to greater disruption of heterochromatin domains. In ∆dim-5 strains, we observed reduced turnover at domain edges and increased turnover within the interior of heterochromatin domains (Figure 4, A and B). Similar increased turnover within the interior of heterochromatin domains was observed in ∆hda-1 strains, but unlike ∆dim-5 strains there were concomitant increases in turnover at domain edges (Figure 4, A and B). Despite these increases, the general profile of histone turnover appeared largely normal in these mutants, perhaps reflecting redundant mechanisms to maintain heterochromatin structure. These results are similar to observations at the S. pombe mating-type locus, where loss of the HDAC Clr3 led to greater increases of histone turnover than from loss of Clr4 and heterochromatin-associated H3K9me (Aygün et al. 2013). The greatest increases in histone turnover were observed in ∆hpo strains. Similar to the case in ∆hda-1 strains, we found increased turnover at domain edges, and histone turnover was much higher overall throughout heterochromatin domains (Figure 4, A and B). That loss of HP1 led to greater destabilization of heterochromatin is puzzling as HP1 is dependent on H3K9me3, and thus DIM-5 catalytic activity, for localization to heterochromatin (Freitag et al. 2004), and DIM-5 is responsible for all H3K9me3 in N. crassa (Tamaru et al. 2003). However, this result is consistent with the observation that HP1 chromodomain mutants retain a low level of DNA methylation, suggesting that HP1 is still recruited to heterochromatin domains despite lacking the ability to bind H3K9me3 (Honda et al. 2016).

Conclusion

We have created a strain that expresses 3xFLAG-tagged histone H3 under the control of a light-inducible promoter to profile histone turnover genome-wide. Inducible tagged histones have been successfully used in the past in multiple organisms to profile histone turnover (Ahmad and Henikoff 2002; Choi et al. 2005; Dion et al. 2007; Rufiange et al. 2007; Aygün et al. 2013; Kraushaar et al. 2013). However, our reporter strain provides greater ease of use and magnitude of expression of H3-3xFLAG, without compromising physiological levels of histone occupancy (Figure 3A). Whereas turnover assays in yeasts, mice, and flies have involved an incubation period for induction and incorporation on the order of hours (Ahmad and Henikoff 2002; Choi et al. 2005; Dion et al. 2007; Rufiange et al. 2007; Aygün et al. 2013; Kraushaar et al. 2013), our strain expresses readily detectible H3-3xFLAG expression and chromatin incorporation as early as 20 min (Figures 1B and Figure 3A), and induction does not require a change of growth medium. Though we used blue light for induction, previous work characterizing the vvd promoter has shown that white light (400–700 nm) from conventional lamps is sufficient (Hurley et al. 2012). The rapid nature of induction in our strain should be useful for profiling histone exchange genome-wide in any context and would be amenable to profile chromatin at regions with quick transcriptional kinetics.